Official government documents reveal new side of defense secretary’s legacy.
Robert S. McNamara, Secretary of Defense from 1961 to 1967, took many secrets with him when he died Monday at 93. But probably no secret was more sensitive politically than the one that would have changed fundamentally the public perception of his role in Vietnam policy had it been become widely known.
The secret was his deliberate deceit of President Lyndon B. Johnson on Aug. 4, 1964 regarding the alleged attack on US warships in the Gulf of Tonkin.
Documents which have been available for decades in the LBJ Library show clearly that McNamara failed to inform Johnson that the U.S. naval task group commander in the Tonkin Gulf, Captain John J. Herrick, had changed his mind about the alleged North Vietnamese torpedo attack on U.S. warships he had reported earlier that day.
By early afternoon Washington time, Herrick had reported to the Commander in Chief Pacific in Honolulu that “freak weather effects” on the ship’s radar had made such an attack questionable. In fact, Herrick was now saying, in a message sent at 1:27 pm Washington time, that no North Vietnamese patrol boats had actually been sighted. Herrick now proposed a “complete evaluation before any further action taken.”
These documents were reviewed by this reporter in researching my book, Perils of Dominance: Imbalance of Power and the Road to War in Vietnam.
McNamara later testified that he had read the message after his return to the Pentagon that afternoon. But he did not immediately call Johnson to tell him that the whole premise of his decision at lunch to approve McNamara’s recommendation for retaliatory air strikes against North Vietnam was now highly questionable.
In fact, no call from McNamara to Johnson was recorded until 3:51 pm — 7 minutes after Johnson called him from his private quarters. Had Johnson been accurately informed about the Herrick message, he might have demanded fuller information before proceeding with a broadening of the war. Johnson had fended off proposals from McNamara and other advisers for a policy of bombing the North on four separate occasions since becoming President.
But when McNamara called Pacific Admiral Grant Sharp shortly after speaking with Johnson, it was not to order a full investigation or to seek more detailed information. In fact, McNamara didn’t even bring up the Herrick report. Instead, he seemed determined to obtain a statement from Sharp that would make it unnecessary to wait for further investigation. “There isn’t any possibility there was no attack, is there?” asked McNamara.
Sharp insisted, however, that the commander on the scene was saying “the situation’s in doubt” and suggested that McNamara “hold this execute” — meaning the strike order to CINCPAC and Seventh Fleet — “until we have a definite indication that this happened….” Sharp said he believed he could get a “definite indication” that the event had occurred within two hours.
But McNamara rejected Sharp’s proposal to wait for confirmation of the attack. Instead he said, “[I]t seems to me we ought to go ahead on that basis: get the pilots briefed, get the planes armed, get everything lined up to go. Continue the execute order in effect, but between now and 6 o’clock get a definite fix and you call me directly.”
McNamara didn’t claim that he had authority from Johnson to make that decision.
After the conversation with Sharp, McNamara didn’t call LBJ to report on what Sharp had told him or what they had agreed on, according to White House phone logs. Instead he went ahead on his own to issue the execute order at 4:49 pm.
The next phone call, which came just one minute after that order was sent, did not come from McNamara but from LBJ. That brief phone conversation, which was not recorded, was followed moments later by a call from McNamara to Johnson in which he the Secretary said the story had already been broken by wire services that a meeting at the White House that night would brief congressional leaders about a second attack on U.S. warships.
McNamara urged Johnson to authorize a statement by the Pentagon about the attack. He’d somehow found the time during the previous hour to draft a statement reaffirming the attack, which he read to Johnson. It said two U.S. warships had been attacked by patrol boats, but that the North Vietnamese boats had been “driven off.” It concluded, “We believe several of the patrol boats were sunk. Details won’t be available till daylight.”
Neither McNamara nor Johnson alluded in that conversation to Admiral Sharp’s seeking confirmatory evidence — a matter that would surely have been on LBJ’s mind if McNamara had told him about it.
The record of phone McNamara-Johnson conversations on the afternoon of Aug. 4, 1964 thus shows a President who was blissfully unaware that the original reports of an attack were now in doubt and that the Commander-in-Chief of Pacific forces was still seeking to obtain confirmation of the attack.
Ultimately, National Security Council documents declassified in 2005 (PDF) would reveal that no attack on US warships had taken place.
It “is not simply that there is a different story as to what happened; it is that no attack happened that night,” they said. “In truth, Hanoi’s navy was engaged in nothing that night but the salvage of two of the boats damaged on August 2.”
Reporter confronted McNamara in 2004
This writer confronted McNamara with that record in a phone conversation with him on Feb. 24, 2004. His response was that telephone calls were not the only way he had to communicate with Johnson and that he could have told Johnson about the military’s unresolved doubts at the National Security Council meeting which took place that night at 6:15 pm.
Unfortunately for McNamara’s alibi, detailed official notes of that Council meeting taken by NSC staffer Bromley Smith, marked “Top Secret Sensitive, For the President’s Eyes Only,” show that McNamara again asserted unequivocally that the attack had indeed taken place.
After USIA Director Carl Rowan asked, “Do we know for a fact that the North Vietnamese provocation took place?” McNamara said, “We will know definitely in the morning.”
When I read those quotes to McNamara over the phone, he suggested that the notes were “not complete.” But McNamara was admitting, in effect, that he did not inform LBJ that afternoon about the Herrick report or about Sharp’s plea to hold off the execute order until confirming evidence had been obtained.
The records of the Tonkin Gulf crisis in the LBJ library also include documentation showing LBJ wanted to get the truth about what McNamara knew and when he knew it.
Even before the Gulf of Tonkin resolution was approved by the Senate Aug. 7, LBJ ordered a full account of the communications between the commanders of U.S. Pacific forces and the Pentagon on Aug. 4 and 5. The requested study was referred to as the “inquiry,” according to a handwritten note on a draft chronology prepared at the Pentagon. It was to be based on the original tapes of all such communications, which were tracked down and transcribed.
McNamara altered transcripts of calls
The clearest evidence that McNamara was afraid of what the inquiry would reveal about his maneuvering on Aug. 4 is that the chronology produced under his personal guidance — where numerous changes were made on previous drafts in McNamara’s own handwriting — deliberately suppressed the most damning words from the transcript of his conversation with Sharp.
Not only did the official Defense Department chronology change the wording of McNamara’s question to Sharp so that it was no longer obviously a leading question, it also failed to mention Sharp’s revelation that Herrick considered the “whole situation” to be “in doubt” and was calling for “daylight recce” — or reconnaissance.
In addition, the McNamara’s chronology portrayed him as agreeing with Sharp that the execute order should be delayed until definite evidence of an attack was obtained. It reports, “McNamara says that even if definite confirmation of an attack is not forthcoming for another 2 hours, an hour would still remain and the execute order could then be issued.”
But McNamara had not said that the executive order could be issued after getting confirmation of an attack. He had said the opposite: “Continue the execute order in effect, but between now and 6 o’clock get a definite fix and you call me directly.” So that crucial sentence was omitted from the chronology.
McNamara did not want LBJ to know that he had rejected Sharp’s proposal to hold the execute order until the situation was clarified and had not even informed him.
There is more evidence in the presidential tapes at the LBJ Library that Johnson believed that McNamara had misinformed him about what had happened in the Tonkin Gulf. Six weeks later, McNamara and then-Secretary of State Dean Rusk went to Johnson with yet another claim that North Vietnamese boats had attacked a U.S. warship in the Tonkin Gulf and again urged a retaliatory bombing of the North.
This time Johnson expressed skepticism and complained about McNamara’s claim of an attack on Aug. 4. “You just came in a few weeks ago and said they’re launching an attack on us — they’re firing at us,” Johnson tells McNamara on the tape recording of the conversation, “and we got through with the firing and concluded maybe they hadn’t fired at all.”
Whether or not Johnson understood the seriousness of McNamara’s deception Aug. 4, he seemed to become even more resistant to McNamara’s views on Vietnam after the incident. That fall, during the presidential election campaign, Johnson began to challenge McNamara for his advocacy of bombing North Vietnam, even referring to his proposal as “your bombing bullshit,” according to accounts given by Assistant Secretary of Defense John McNaughton to Daniel Ellsberg, his then-assistant (who later leaked what would be known as the Pentagon papers).
McNamara’s deception was one of many maneuvers aimed at pulling Johnson into an escalated war in Vietnam. But it is perhaps the only one in which McNamara’s role shifted from tough bureaucratic in-fighting to usurping presidential authority, in effect, on an issue involving the use of military force.
On the following audio file from The Lyndon Baines Johnson Library, Johnson can be heard telling McNamara that he had misinformed him about the alleged attack:
A pdf file of the McNamara/Sharp transcript can be accessed at this link.
A pdf file of the chronology of events that McNamara prepared for Johnson can be accessed at this link.
Gareth Porter is an investigative journalist and historian who covered the Vietnam War as Saigon Bureau Chief for Dispatch News Service International in 1970-71 and taught international studies at the City College of New York and American University from 1982 to 1990. His book Perils of Dominance: Imbalance of Power and the Road to War in Vietnam, documents US policymaking on Vietnam during the Eisenhower, Kennedy and Johnson administrations. He has a doctorate in Southeast Asian politics from Cornell University, and currently covers diplomatic and military affairs for Inter Press Service.